Bahare Fatemi
@BahareFatemi
Followers
2K
Following
1K
Statuses
458
Research Scientist @GoogleAI, PhD from @UBC_CS, Before: @MetaAI, @element_ai, and @borealisai.
Montreal, Quebec
Joined June 2017
Think your LLM is a time wizard? 🧙 One might think they answer temporal questions through reasoning, but under the hood they may solely rely on their parametric knowledge 🤯 Our new benchmark, ToT, is specifically designed to measure the temporal reasoning capabilities of LLMs.
4
10
65
RT @mirrokni: The paper introduces a representational hierarchy categorizing graph problems by complexity, and characterized transformers'…
0
3
0
RT @mirrokni: Check out the blog post about our NeurIPS'24 paper on graphs and reasoning. We explored the theoretical and practical underpi…
0
42
0
RT @ReyhaneAskari: "Diffusion-Based In-painting of Corrupted Spectrogram" We propose a series of solutions to remove radio frequency inter…
0
1
0
RT @ReyhaneAskari: "Deliberate practice with Synthetic data" We propose a new framework to modify  text-to-image models to generates synthe…
0
2
0
RT @ReyhaneAskari: "On Improved Conditioning Mechanisms and Pre-training Strategies for Diffusion Models" This paper explores the state sp…
0
1
0
RT @ReyhaneAskari: Excited to be at #NeurIPS this week! 🚨I'm wrapping up my postdoc at FAIR and I'm on the job market looking for RS posit…
0
29
0
Special thanks to my coauthors: @tsitsulin_, @kazemi_sm, @jhalcrow, @phanein, @mirrokni, @NishanthDikkala, @ankit_s_anand, @hardy_qr, @ivakshi_s, @mariojfritz, @tangjiliang and more.
0
0
8
RT @kazemi_sm: [4] Understanding transformer reasoning capabilities via graph algorithms Collaborated with @BahareFatemi, @tsitsulin_, @ph…
0
1
0
RT @kazemi_sm: [3] Generative Verifiers: Reward Modeling as Next-Token Prediction Led by @LunjunZhang in collaboration with @arianTBD, @hb…
0
1
0
RT @kazemi_sm: [2] Smaller, Weaker, Yet Better: Training LLM Reasoners via Compute-Optimal Sampling Led by @hbXNov in collaboration with @…
0
1
0
RT @kazemi_sm: [1] ReMI: A Dataset for Reasoning with Multiple Images Work done with @NishanthDikkala, @ankit_s_anand, @hardy_qr, @BahareF…
0
2
0
RT @kazemi_sm: Sad to miss #NeurIPS this year! But you can still catch four of our latest and greatest works on multi-image reasoning, com…
0
5
0